1. Field of the Invention
This invention relates to decoding apparatus of a digital video signal compressed by highly efficient coding.
2. Description of the Prior Art
When a digital video signal is transmitted or recorded/reproduced, highly efficient coding for compressing data is frequently employed. The highly efficient coding compresses the transmission data by using spatial correlation and time correlation of an image. For highly efficient coding of a video signal, intraframe encoding and interframe encoding are known. A difference lies in that data to be encoded is intraframe picture data or a difference between two consecutive frames in time between these encoding.
For an example of the highly efficient coding, transform coding is known. The transform coding, particularly two-dimensional ones, divides picture data into blocks of (8.times.8) picture elements, for example, and performs orthogonal transformation for each block. Transformation components (called coefficient data) are their direct current components and high frequency components. Generally, since the direct current components are large and the high frequency components are small, the bit number as a whole can be reduced by allocating an appropriate bit number to each coefficient. Recently, DCT (Discrete Cosine Transform) has particularly been drawing attention.
In the case of the intraframe coding, picture data of one frame is segmented into two-dimensional blocks such as (8.times.8), and each block is DCT-encoded. Further, by DCT-encoding a difference between two consecutive frames in terms of time, interframe coding is achieved. In general, the intraframe coding does not need a frame memory to thereby reduce the hardware size but the data compression rate is low. On the other hand, the interframe coding requires a frame memory but the data compression rate is high. In view of these points, when highly efficient coding is adopted in a digital VTR, an encoding method to be employed is selected in consideration of the tape consumption amount at the time of recording. Also, it is considered that in a camera-integrated VTR, the interframe coding is adopted, while in a stationary VTR, the intraframe coding is used.
However, when there is a difference between the highly efficient coding methods in two kinds of digital VTRs, a picture recorded by one VTR cannot be reproduced by the other.